🧪 add long context test

[joerunde]

Description

This adds a failing test case for handling batch size 4 at 4k context length with continuous batching.

This tests the eager backend against the sendnn backend, which is currently failing on quay.io/ibm-aiu/spyre-base:2025_07_18-amd64

[github-actions]

👋 Hi! Thank you for contributing to vLLM support on Spyre.
Just a reminder: Make sure that your code passes all the linting checks, otherwise your PR won't be able to be merged. To do so, first install the linting requirements, then run format.sh and commit the changes. This can be done with uv directly:

uv sync --frozen --group lint --active --inexact

Or this can be done with pip:

uv pip compile --group lint > requirements-lint.txt
pip install -r requirements-lint.txt
bash format.sh

Now you are good to go 🚀

[yannicks1]

lgtm